Astronomy Dataset for Multimodal Models

# Space Vision Dataset (Multimodal) ## Overview The **Space Vision Dataset** is a multimodal dataset consisting of images paired with descriptive captions. It includes a diverse collection of space-related imagery such as planetary views, telescopes, galaxies, and Mars rover scenes. This dataset is designed for tasks like: * Image Captioning* Vision-Language Modeling (VLM)* Multimodal Retrieval* Contrastive Learning --- ## Dataset Structure Each sample contains: * `image_id` — Unique identifier for the image* `image` — Image file* `text` — Natural language description of the image ### Example ```json{ "image_id": "001", "text": "A satellite image of Earth showing the curvature and scattered clouds over the Atlantic Ocean.", "image": "images/001.jpg"}``` --- ## Categories The dataset includes a variety of space-related themes: * Earth (satellite & orbital views)* Mars (surface, orbit, and geology)* Space Telescopes (Hubble imagery)* Milky Way (astronomical observations)* Mars Rovers (exploration scenes) --- ## File Organization ```dataset/├── dataset.json└── images/ ├── 001.jpg ├── 002.jpg └── ...``` --- ## Usage Load with 🤗 Datasets: ```pythonfrom datasets import load_dataset dataset = load_dataset("AIOmarRehan/space-multimodal-dataset")``` --- ## Applications This dataset can be used for: * Training image captioning models (e.g., BLIP, LLaVA)* Fine-tuning multimodal LLMs* Benchmarking vision-language tasks* Educational and research purposes --- ## License Specify your license here (e.g., MIT, Apache 2.0, or custom). --- ## Acknowledgements This dataset was created for experimentation and research in multimodal AI systems. --- ## Contact For questions or contributions, feel free to open an issue or reach out.